def test_replace_all_errors():
s = fake.session()
project = fake.mastering_project()
transforms = fake.transforms()
with pytest.raises(HTTPError):
tc.transformations.replace_all(s, project, transforms)
def test_from_project():
s = fake.session()
project = fake.mastering_project()
unified_dataset = tc.dataset.unified.from_project(s, project)
assert unified_dataset.name == "dataset 1 name"
assert unified_dataset.description == "dataset 1 description"
assert unified_dataset.key_attribute_names == ("tamr_id",)
def test_replace_all():
s = fake.session()
project = fake.mastering_project()
transforms = fake.transforms()
transforms.unified_scope.append("//extra TX")
transforms.input_scope.pop(1)
tc.transformations.replace_all(s, project, transforms)
def test_from_project_dataset_not_found():
s = fake.session()
instance = fake.instance()
project = fake.mastering_project()
url = tc.URL(path="projects/1/unifiedDataset")
responses.add(responses.GET, str(url), status=404)
with pytest.raises(tc.dataset.unified.NotFound):
tc.dataset.unified.from_project(s, instance, project)
def test_update_cluster_results_async():
s = fake.session()
project = fake.mastering_project()
op = tc.mastering._update_cluster_results_async(s, project)
assert op.type == "SPARK"
assert op.description == "Materialize views to Elastic"
assert op.status == {
"state": "PENDING",
"startTime": "",
"endTime": "",
"message": "Job has not yet been submitted to Spark",
}
def test_estimate_pairs_async():
s = fake.session()
project = fake.mastering_project()
op = tc.mastering._estimate_pairs_async(s, project)
assert op.type == "SPARK"
assert op.description == "operation 1 description"
assert op.status == {
"state": "PENDING",
"startTime": "",
"endTime": "",
"message": "Job has not yet been submitted to Spark",
}
def test_from_project():
s = fake.session()
instance = fake.instance()
project = fake.mastering_project()
dataset_json = utils.load_json("dataset.json")
url = tc.URL(path="projects/1/unifiedDataset")
responses.add(responses.GET, str(url), json=dataset_json)
unified_dataset = tc.dataset.unified.from_project(s, instance, project)
assert unified_dataset.name == "dataset 1 name"
assert unified_dataset.description == "dataset 1 description"
assert unified_dataset.key_attribute_names == ("tamr_id", )
def test_get_all():
s = fake.session()
project = fake.mastering_project()
transforms = tc.transformations.get_all(s, project)
assert len(transforms.input_scope) == 2
assert len(transforms.unified_scope) == 1
assert len(transforms.input_scope[0].datasets) == 0
assert transforms.input_scope[0].transformation == "SELECT *, 1 as one;"
assert len(transforms.input_scope[1].datasets) == 1
assert transforms.input_scope[1].datasets[0].name == "dataset 1 name"
assert transforms.input_scope[1].transformation == "SELECT *, 2 as two;"
assert transforms.unified_scope[0] == "//Comment\nSELECT *;"
def test_from_project_dataset_not_found():
s = fake.session()
project = fake.mastering_project()
with pytest.raises(tc.dataset.unified.NotFound):
tc.dataset.unified.from_project(s, project)